Private companies aren’t the only ones eyeing the Mid-Columbia for its potential as a site for data centers and development of artificial intelligence.

The U.S. Department of Energy (DOE) recently requested data center and energy developers, as well as the general public, provide input on how the agency could support AI infrastructure at 16 proposed sites, including Pacific Northwest National Laboratory.

Calling the global race for AI dominance “the next Manhattan project,” U.S. Energy Secretary Chris Wright said the agency’s national labs offer expertise in developing AI tools and related disciplines that give the nation an edge in being a leader in AI.

“The Department of Energy is taking important steps to leverage our domestic resources to power the AI revolution, while continuing to deliver affordable, reliable and secure energy to the American people,” Wright said in a statement.

The agency’s Request for Information, or RFI, indicates that a variety of input is sought, including but not limited to industry interest in the proposed DOE sites, potential designs and power needs, economic and environmental impacts and possible challenges.

In detailing PNNL’s potential as a site, the RFI cites the lab’s dry climate, minimal risk of natural disaster and potential for power availability via small modular reactors, or SMRs, currently being developed by Energy Northwest and Amazon. It also highlighted its “access to the highly skilled labor needed to construct and operate the power generation facilities; and to quickly build the proposed frontier AI data center, benefiting from construction costs that are significantly lower than the national average.”

“These are among the reasons that major hyper-scalers have chosen eastern Washington for their large data centers,” the document states.

The lab already has a track record in leveraging and advancing AI in its research. In late 2023, its leaders launched the Center for AI@PNNL, combining fundamental science and computing to accelerate the rate of scientific discovery required to address many of the world’s most urgent problems.  

“The Trump Administration has demonstrated that AI is a top priority, and in keeping with this it is exciting to see the Department of Energy pursue innovative approaches to artificial intelligence infrastructure development – including a potential presence in the Tri-Cities area that many in the community are eager to explore,” according to a PNNL statement.

And the RFI does have local leaders’ attention. The Energy Forward Alliance, part of the Tri-City Development Council, is coordinating with regional partners to review the document and determine how a unified response could ensure the region is part of DOE’s efforts.

“The Energy Forward Alliance was created to help unlock new economic opportunities tied to collaborative innovation and clean energy in the Tri-Cities and greater Mid-Columbia region,” said Sean O’Brien, EFA’s executive director, in a statement. “We view the Department of Energy’s interest in expanding AI infrastructure as a promising opportunity to advance that vision, with the potential to build upon our region’s scientific expertise, strategic location, robust infrastructure, skilled workforce, and longstanding leadership in energy and technology.”

The Tri-Cities has already seen its share of data center interest. Earlier this year the city of West Richland, at the request of a large property owner, updated its zoning to allow the construction of data centers in specific areas. 

A yet-unidentified American company is in discussions with the Port of Walla Walla to potentially buy 500 acres of port land near Wallula to establish a data center campus that could create hundreds of jobs and inject $5 billion into the region’s economy. 

But there are potential obstacles to AI infrastructure becoming a significant industry in the region. A recent report from CBRE Inc., a global commercial real estate brokerage, showed that in the past year, even as the region added 84 megawatts of power capacity for data centers, less than half a megawatt remains unused, indicating high demand.